The present proposal is based on our previous studies which demonstrated that peripheral BFU-E in sickle cell anemia patients varied in their characteristics according to the peripheral percent of HbF. Specifically, low HbF patients (< 9%), have a higher number of BFU-E, with a high proportion of these BFU-E in active cycling and adherent mononuclear cells obtained from these patients constitutively produce BPA. More recent work has demonstrated that GM-CSF is either the sole or most important component of this BPA activity. In addition, preliminary data demonstrates that light density mononuclear cells harvested from peripheral blood of high HbF SS patients (> 9%) produce negative growth factors for BFU-E. Finally, using delayed addition of EpO we have detected heterogeneity in the intrinsic characteristics of BFU-E in SS disease: some BFU-E are sensitive to EpO alone, others to GM-CSF and IL-3 and yet others to IL-3 alone. The heterogeneity also varies with peripheral HbF levels. The SPECIFIC AIMS of this proposal are the following: What are the positive growth factors involved in BPA derived from the adherent cells that affect BFU-E in low HbF SS patients? Besides GM-CSF, is IL-3 involved? What are the negative growth factors involved in the regulation of BFU-E in sickle cell anemia patients with high HbF?; Are GM-CSF, and other growth factors, found in detectable levels in blood of sickle cell anemia patients? Is this related with HB F levels?; Using the strategy of delayed addition of EpO to distinguish populations of BFU-E according to their differential sensitivity to EpO, GM-CSF and IL-3, ask: Do patients with sickle cell anemia have populations of BFU-E with diverse sensitivities to growth factors? If so, how is the distribution of growth factor sensitivities correlated with HbF and other hematological parameters of SS disease?